A previously unknown metabolic pathway for the utilization of l-galactose was discovered in a prevalent gut bacterium, Bacteroides vulgatus. The new pathway consists of three previously uncharacterized enzymes that were found to be responsible for the conversion of l-galactose to d-tagaturonate. Bvu0219 (l-galactose dehydrogenase) was determined to oxidize l-galactose to l-galactono-1,5-lactone with kcat and kcat/Km values of 21 s(-1) and 2.0 × 10(5) M(-1) s(-1), respectively. The kinetic product of Bvu0219 is rapidly converted nonenzymatically to the thermodynamically more stable l-galactono-1,4-lactone. Bvu0220 (l-galactono-1,5-lactonase) hydrolyzes both the kinetic and thermodynamic products of Bvu0219 to l-galactonate. However, l-galactono-1,5-lactone is estimated to be hydrolyzed 300-fold faster than its thermodynamically more stable counterpart, l-galactono-1,4-lactone. In the final step of this pathway, Bvu0222 (l-galactonate dehydrogenase) oxidizes l-galactonate to d-tagaturonate with kcat and kcat/Km values of 0.6 s(-1) and 1.7 × 10(4) M(-1) s(-1), respectively. In the reverse direction, d-tagaturonate is reduced to l-galactonate with values of kcat and kcat/Km of 90 s(-1) and 1.6 × 10(5) M(-1) s(-1), respectively. d-Tagaturonate is subsequently converted to d-glyceraldehyde and pyruvate through enzymes encoded within the degradation pathway for d-glucuronate and d-galacturonate.
A previously unknown metabolic pathway for the utilization of l-galactose was discovered in a prevalent gut bacterium, Bacteroides vulgatus. The new pathway consists of three previously uncharacterized enzymes that were found to be responsible for the conversion of l-galactose to d-tagaturonate. Bvu0219 (l-galactose dehydrogenase) was determined to oxidize l-galactose to l-galactono-1,5-lactone with kcat and kcat/Km values of 21 s(-1) and 2.0 × 10(5) M(-1) s(-1), respectively. The kinetic product of Bvu0219 is rapidly converted nonenzymatically to the thermodynamically more stable l-galactono-1,4-lactone. Bvu0220 (l-galactono-1,5-lactonase) hydrolyzes both the kinetic and thermodynamic products of Bvu0219 to l-galactonate. However, l-galactono-1,5-lactone is estimated to be hydrolyzed 300-fold faster than its thermodynamically more stable counterpart, l-galactono-1,4-lactone. In the final step of this pathway, Bvu0222 (l-galactonate dehydrogenase) oxidizes l-galactonate to d-tagaturonate with kcat and kcat/Km values of 0.6 s(-1) and 1.7 × 10(4) M(-1) s(-1), respectively. In the reverse direction, d-tagaturonate is reduced to l-galactonate with values of kcat and kcat/Km of 90 s(-1) and 1.6 × 10(5) M(-1) s(-1), respectively. d-Tagaturonate is subsequently converted to d-glyceraldehyde and pyruvate through enzymes encoded within the degradation pathway for d-glucuronate and d-galacturonate.
Most of the
microorganisms that
colonize the human body reside within the gastrointestinal tract and
are collectively known as gut microbiota. These microorganisms are
of particular interest for the role they play in metabolism and human
health.[1] While the human–microbiota
relationship is predominantly commensal, there are reports implicating
these microbes in a wide range of pathologies, including inflammatory
bowel disease and colitis.[2] One important
role for these organisms is the degradation of poorly digestible plant
polysaccharides such as pectin, xylan, and cellulose.[3]Bacteroides species are prevalent in the
microbiota, and these organisms have been implicated as being partly
responsible for the degradation of these polysaccharides. The various
species of Bacteroides encode an unusually high number
of glycoside hydrolases and polysaccharide lyases. Bacteroides
vulgatus is thought to encode the largest number of enzymes
targeting the degradation of pectin.[3−5]A cluster of genes
that is likely responsible for an unknown metabolic
pathway for the catabolism of carbohydrates in B. vulgatus is presented in Figure 1. Of the six genes
identified in this cluster, four of them likely function to code for
enzymes that utilize carbohydrate-based substrates. On the basis of
amino acid sequence identity to proteins of known function, Bvu0217
is a putative glycoside hydrolase from cog3669, Bvu0219 is a NAD+/NADP+-dependent oxidoreductase from cog0667, Bvu0220
is a lactonase associated with cog3618, and Bvu0222 is a Zn2+-dependent dehydrogenase from cog1063. The two remaining proteins,
Bvu0218 and Bvu0221, appear to be a transcriptional regulator and
a sugar permease from cog0738, respectively. A potential metabolic
pathway for these enzymes involves the initial hydrolysis of a more
complex carbohydrate substrate (glycohydrolase), oxidation of the
aldose product to a lactone (dehydrogenase), hydrolysis of the lactone
to an acid sugar (lactonase), and the subsequent oxidation of the
acid sugar (dehydrogenase).
Figure 1
Genomic neighborhood of Bvu0220. Genes colored
in maroon have been
functionally characterized by this investigation and are as follows:
Bvu0219 is l-galactose dehydrogenase, Bvu0220 is l-galactono-1,5-lactonase, and Bvu0222 is l-galactonate dehydrogenase.
The genes in white are of unknown function. Bvu0217 is a putative
glycoside hydrolase, Bvu0218 is a putative transcription regulator,
and Bvu0221 is a putative sugar permease.
Genomic neighborhood of Bvu0220. Genes colored
in maroon have been
functionally characterized by this investigation and are as follows:
Bvu0219 is l-galactose dehydrogenase, Bvu0220 is l-galactono-1,5-lactonase, and Bvu0222 is l-galactonate dehydrogenase.
The genes in white are of unknown function. Bvu0217 is a putative
glycoside hydrolase, Bvu0218 is a putative transcription regulator,
and Bvu0221 is a putative sugar permease.Of the enzymes contained within this putative pathway for
the catabolism
of an unknown carbohydrate, we were particularly interested in the
function of Bvu0220, a member of the amidohydrolase superfamily (AHS).
The AHS is a diverse superfamily of enzymes that catalyzes a wide
variety of reactions including the hydrolysis of amide or ester bonds,
hydration, decarboxylation, and isomerization reactions.[6−10] Most of the proteins in the AHS exhibit a distorted (β/α)8-barrel topology, with the active site located at the C-terminal
end of the β-barrel. Nearly all of the structurally characterized
enzymes contained within the AHS bind one to three divalent cations,
which are ligated to the protein via complexation with a common set
of conserved amino acid residues. However, the structurally characterized
lactonases from the AHS that are contained within cog3618 do not function
with a divalent cation bound in the active site. The known lactonases
from cog3618 include 2-pyrone-4,6-dicarboxylate lactonase (LigI),
4-sulfomuconolactonase (4-SML), l-rhamnonate-1,4-lactonase,
and l-fucono-1,5-lactonase (FucL).[11−13]A sequence
similarity network (SSN) for cog3618 calculated at a
BLAST E-value of 10–30 is provided in Figure 2. Each node within the sequence similarity network
represents a unique protein contained within cog3618, which are connected
to each other if they are related by an E-value of 10–30 or less. At this stringency level, cog3618 is subdivided into two
primary groups, arbitrarily designated as Class I and Class II. Most
of the enzymes within Class I are encoded adjacent to genes with an
apparent relationship to carbohydrate metabolism.[13] At this level of stringency, Bvu0220 is not associated
with either Class I or Class II enzymes. Here we identify a novel
metabolic pathway for the metabolism of l-galactose to d-tagaturonate. l-Galactose is first oxidized to l-galactono-1,5-lactone and then hydrolyzed to l-galactonate.
This product is subsequently oxidized to d-tagaturonate,
which is further metabolized to pyruvate and d-glyceraldehyde-3-phosphate.
Figure 2
Sequence
similarity network for cog3618 at an E-value of 10–30 where each node represents a protein and an edge
represents an E-value between two proteins of 10–30 or smaller. The triangular nodes represent an enzyme with experimentally
verified function. Nodes are color-coded as follows: LigI (red), 4-SML
(orange), l-rhamnonolactonase (blue), FucL (green), and Bvu0220
(yellow).
Sequence
similarity network for cog3618 at an E-value of 10–30 where each node represents a protein and an edge
represents an E-value between two proteins of 10–30 or smaller. The triangular nodes represent an enzyme with experimentally
verified function. Nodes are color-coded as follows: LigI (red), 4-SML
(orange), l-rhamnonolactonase (blue), FucL (green), and Bvu0220
(yellow).
Materials and Methods
Materials
All
chemicals and buffers were purchased
from Sigma-Aldrich, unless otherwise specified. Sugar lactones that
were not commercially available were synthesized according to standard
published procedures with the exception of 4-deoxy-l-fucono-1,5-lactone,
which was enzymatically synthesized.[13,14] The noncommercial
lactones included the following: l-fucono-1,4-lactone, d-arabinono-1,4-lactone, l-xylono-1,4-lactone, l-mannono-1,4-lactone, d-talono-1,4-lactone, d-allono-1,4-lactone, l-rhamnono-1,4-lactone, d-lyxono-1,4-lactone, l-lyxono-1,4-lactone, l-arabinono-1,4-lactone, d-xylono-1,4-lactone, l-mannono-1,5-lactone, l-rhamnono-1,5-lactone, and 4-deoxy-l-fucono-1,5-lactone.
The sugar lactones that were obtained from either CarboSynth or ChromaDex
included the following compounds: l-galactono-1,4-lactone, d-mannono-1,4-lactone, d-galactono-1,4-lactone, d-ribono-1,4-lactone, d-glucurono-6,3-lactone, d-erythronolactone, and l-glucono-1,5-lactone. Aldose
sugars were obtained from either Sigma-Aldrich or Carbosynth and included l-fucose, l-galactose, l-glucose, d-altrose, d-arabinose, l-xylose, l-rhamnose, d-mannose, l-allose, d-talose, l-talose, d-allose, d-galactose, l-mannose, d-gulose, d-glucose, l-arabinose, l-ribose, d-lyxose, l-lyxose, d-xylose, d-ribose,
and 4-deoxy-l-fucose.
Cloning, Expression and
Purification of Bvu0220 from B. vulgatus
The gene for Bvu0220 (gi|150002825;
UniProt: A6KWY3) was amplified from B. vulgatus ATCC
8482 genomic DNA using 5′-TACTTCCAATCCATGATGGATTTTAACATAATAGATGCACACTCC-3′
as the forward primer and 5′-TATCCACCTTTACTGTTATTCTGACATGTTTTTTATGTAAGGAAGAGATATG-3′
as the reverse primer. PCR was performed using KOD Extreme DNA Polymerase
(Novagen). The conditions were 2 min at 95 °C, followed
by 40 cycles of 20 s at 95 °C, 20 s at 66 °C,
and 20 s at 72 °C. The amplified fragment was cloned into the
N-terminal TEV cleavable 6x-His-tag containing vector, pNIC28-BSA,
by ligation-independent cloning.[15]The recombinant plasmid containing Bvu0220 was transformed into BL21
(DE3) cells (Novagen), via electroporation. Five milliliter cultures
of LB medium supplemented with 50 μg/mL kanamycin were inoculated
with a single colony and grown overnight at 37 °C. Overnight
cultures were used to inoculate 1 L of LB medium (50 μg/mL kanamycin)
and incubated at 37 °C until an OD600 of 0.4–0.6
was achieved. Gene expression was induced by the addition of 0.1 mM
isopropyl β-thiogalactoside (IPTG) at ambient temperature for
18 h. Cells were harvested via centrifugation at 8000 rpm at 4 °C.
Cell pellets were resuspended in purification buffer (50 mM HEPES,
500 mM NaCl, 40 mM imidazole, 0.1 mg/mL phenylmethanesulfonyl fluoride,
pH 7.5) and lysed via multiple rounds of sonication. Nucleic acids
were removed from the cell lysate with the addition of 2% (w/v) protamine
sulfate over 30 min at 4 °C, followed by centrifugation (8000
rpm). The resulting supernatant was applied to a 5 mL HisTrap HP (GE
Healthcare) nickel affinity column. Protein was eluted from the nickel
affinity column with 50 mM HEPES (pH 7.5), 0.5 M NaCl, and 500 mM
imidazole over a gradient of 23 column volumes.
Cloning, Expression,
and Purification of Bvu0219 and Bvu0222
from B. vulgatus
The gene for Bvu0219 (gi|150002825;
UniProt: A6KWY2) was cloned from B. vulgatus ATCC
8482 with the primer pair 5′-ATGAATTACAATGAAATGGGAAAACCGGTATGCGGGTT-3′
and 5′-TGAATTAGCCCAACTGATACGCATTTGTTTTCCGATAA-3′.
Restriction sites for Nhe1 and Xho1 were inserted into the forward
and reverse primers. The PCR product was purified with a Promega PCR
cleanup system and subsequently digested with Nhe1 and HindIII. The resulting DNA product was ligated into a pET-24a(+) vector
(Novagen). The ligation reaction mixture was transformed into BL21(DE3)
cells via electroporation. A single colony containing the plasmid
of interest was used to inoculate 5 mL overnight cultures of LB (50
μg/mL kanamycin) and incubated at 37 °C. One liter cultures
of LB containing 50 μg/mL kanamycin were inoculated with a 5
mL overnight culture and then grown at 37 °C until an OD600 of 0.4–0.6 was obtained. The cultures were transferred
to ambient temperature, and gene expression was initiated with the
addition of 1.0 mM isopropyl IPTG. Cells were harvested after 18 h
via centrifugation at 4 °C and suspended in 50 mM HEPES (pH 7.5),
0.25 mM NaCl, and 0.1 mg/mL of PMSF. Resuspended cells were lysed
via multiple rounds of sonication, and cell debris was removed via
centrifugation at 4 °C. Nucleic acids were removed from the supernatant
solution with the dropwise addition of 2% (w/v) protamine sulfate
at 4 °C. Protamine sulfate-bound DNA was removed by centrifugation.
Ammonium sulfate was added between 30 and 50% saturation, and the
precipitated protein was isolated by centrifugation. The protein was
resuspended in 20 mM HEPES (pH 7.5) and then applied to a High Load
26/60 Superdex 200 gel filtration column (GE Healthcare). Fractions
containing the protein of expected size were pooled and purified further
by ion exchange chromatography with a ResourceQ column (6 mL) at pH
7.5.The gene coding for Bvu0222 (gi|149931251; UniProt: A6KWY5)
was determined to contain 19 rare codons using the Rare Codon Calculator
(RaCC) (nihserver.mbi.ucla.edu/RACC/).[16] A codon-optimized gene was purchased from GenScript, which was ligated
into a pUC57 cloning vector containing Nde1 and HindIII restriction sites. The plasmid harboring the gene of interest
was digested with Nde1 and HindIII at 37 °C
for 3 h. The resulting gene was ligated into pET-32a (+) (Novagen)
and transformed into BL21(DE3) cells via electroporation. The purification
of Bvu0222 was conducted in manner similar to that described for Bvu0219.
The isolation of uronate isomerase (URI) from Escherichia
coli and l-fucose dehydrogenase (FucD) from Burkholderia multivorans (BmulJ_04919) was conducted as
previously described.[13,17]
Measurement of Enzyme Activity
The enzymatic hydrolysis
of lactones by Bvu0220 was monitored with a SpectraMax340 UV–visible
using a colorimetric pH indicator assay at 30 °C, as previously
described.[13] Protons released during the
hydrolysis of lactones were measured using the pH indicator bromothymol
blue. Reaction mixtures contained 2.5 mM MOPS (pH 7.1), 0.2 M NaCl,
0–0.5 mM lactone, and 0.1 mM bromothymol blue in a final volume
of 250 μL. The final concentration of DMSO was 1%. Changes in
absorbance at 616 nm (ε = 1135 M–1 cm–1) were monitored in 96-well plates. The background
rate arising from acidification of the reaction mixture by atmospheric
CO2 was subtracted from the initial rates.The dehydrogenase
activities of Bvu0222 and Bvu0219 were monitored by following the
reduction of NAD+ or NADP+ at 340 nm (ε
= 6220 M–1 cm–1) at 30 °C
with a SpectraMax340 UV–visible spectrophotometer. Reaction
assays for Bvu0219 were performed in 50 mM BICINE buffer (pH 8.0),
varying concentrations of substrate, and 0.5 mM NADP+.
Reaction assays for Bvu0222 were conducted in 50 mM phosphate buffer
at pH 7.5. Kinetic constants for NADP+ were determined
with 200 μM substrate and variable concentrations of NADP+ (0–1.5 mM).The reductase activity of Bvu0222
was monitored using a coupled
assay system. The coupling system consisted of 8.0 μM URI, d-galacturonate (0–4 mM), 50 mM phosphate buffer (pH
7.0), NADH (1.0 mM), and Bvu0222. The reactions were allowed to incubate
for 15 min in the absence of Bvu0222 at 30 °C to allow for the
URI catalyzed isomerization of d-galacturonate to d-tagaturonate. Reactions were initiated by the introduction of Bvu0222
and monitored through the oxidation of NADH at 340 nm at 30 °C,
with a SpectraMax340 UV–visible spectrophotometer.
Data Analysis
The kinetic constants were determined
from a fit of the initial velocity data to eq 1 using SigmaPlot 9, where v is the initial velocity, Et is the total enzyme concentration, kcat is the turnover number, [A] is the substrate
concentration, and Km is the Michaelis
constant.
Metal Analysis
The metal content
of Bvu0220 and Bvu0222
was determined using an Elan DRC II ICP-MS instrument as previously
described.[18] Protein samples for ICP-MS
analysis were digested with HNO3 and then refluxed for
30 min.[19] All buffers were passed through
a column of Chelex 100 (Bio-Rad) to remove trace metal contamination.
EDTA and 1,10-phenanthroline (1.0 mM) were incubated with 1.0 μM
Bvu0220 in 50 mM buffer at various pH values ranging from 6 to 10
to remove divalent metal ions, as previously described.[13] The following buffers were used: CHES (pH 6.0),
HEPES (pH 7.0), BICINE (pH 8.0), and CHES (pH 9.0 and 10.0). The effect
of added divalent cations on the catalytic activity of Bvu0220 was
determined by supplementing Mn2+, Zn2+, Co2+, Cu2+, or Ni2+ (0–500 μM)
directly to the assay mixtures. The purified enzyme was also incubated
with 50–500 mol equiv of these divalent cations for 24 h at
4 °C in 50 mM HEPES (pH 7.5) and subsequently assayed for catalytic
activity.
Identification of Reaction Product of l-Galactose Oxidation
Catalyzed by Bvu0219
The product of the reaction catalyzed
by Bvu0219 was determined by 13C NMR spectroscopy using
a Bruker Avance III 500 MHz spectrometer equipped with a 5 mm HCN
cryoprobe. The reaction sample contained 250 mM phosphate (pH 5.8),
10% D2O, 10 mM NAD+, 10 mM [13C-1]-l-galactose (Omicron Biochemicals) and 10–50 μM
Bvu0219 in a volume of 250 μL. The reaction was initiated by
the addition of Bvu0219 and terminated at 10 min by removal of the
enzyme by passage through a 3 kDa cutoff VWR centrifugal filter. The
pH of the sample was adjusted to 7.0 to facilitate the nonenzymatic
transformation of l-galactono-1,5-lactone to the thermodynamically
more stable l-galactono-1,4-lactone.
Hydrolysis of l-Galactono-1,5-Lactone by Bvu0220
l-Galactono-1,5-lactone
was identified as the preferred
substrate for Bvu0220 via 13C NMR spectroscopy. The reaction
was initiated by the addition of 50 μM Bvu0219 to a solution
containing 250 mM phosphate (pH 5.8), 10% D2O, 10 mM NAD+, and 10 mM [13C-1]-l-galactose. The reaction
was monitored by 13C NMR spectroscopy until the resonances
for the 1,4- and 1,5-lactones of l-galactose were approximately
equal. At this point, 1.0 μM Bvu0220 was added, and the 13C NMR spectra were collected as a function of time.
Enzymatic
Synthesis of Acid Sugars and Lactones
The
acid sugars of l-galactose and l-glucose were synthesized
enzymatically from l-galactono-1,4-lactone and l-glucono-1,5-lactone using Bvu0220 as the catalyst. Reaction conditions
were as follows: 50 mM lactone, 50 mM carbonate buffer (pH 9.5), and
20 μM Bvu0220 were incubated together overnight. 4-Deoxy-l-fucono-1,5-lactone was synthesized enzymatically from 4-deoxy-l-fucose using FucD as previously described.[13] The reaction was conducted in 50 mM phosphate buffer (pH
7.0) containing 15 mM 4-deoxy-l-fucose, and 15 mM NAD+ in a final volume of 1.0 mL for 20 min.
Sequence Similarity
Network for cog3618
Approximately
2000 proteins belonging to cog3618 were identified from the NCBI protein
database using the query “cog3618”. The proteins within
cog3618 were subjected to an all-by-all BLAST at a specified E-value (10–50, 10–60, etc.) using the NCBI standalone BLAST program. The BLAST files
were opened and visualized in the similarity network program Cytoscape.[20,21]
Multiple Sequence Alignment
FASTA sequence files were
obtained from the NCBI Protein database for Bvu0220, LigI, and FucL.
Multiple sequence alignments were prepared using JalView 2.8.0.[22]
Results
Functional Characterization
of Bvu0220
The gene encoding
Bvu0220 from B. vulgatus was cloned and expressed
in E. coli, and the protein purified to homogeneity.
The as-purified Bvu0220 contained ∼0.3 mol equiv of Mn2+, 0.1 mol equiv of Fe2+, and 0.3 mol equiv of
Ni2+. The metal ions were removed by incubation of the
enzyme with EDTA and o-phenanthroline overnight.
Since other enzymes from cog3618 have been shown previously to hydrolyze
γ- and δ-lactones of acid sugars, Bvu0220 was incubated
with a small library of 25 acid sugar lactones as a preliminary test
of catalytic activity. Of the compounds tested, hydrolysis was detected
only with l-glucono-1,5-lactone (1), l-galactono-1,4-lactone (2), l-fucono-1,4-lactone
(3), d-arabino-1,4-lactone (4),
and l-xylono-1,4-lactone (5). The structures
of these compounds are illustrated in Scheme 1. The activity of Bvu0220 was not affected by the addition of chelating
agents, or the addition of Mn2+, Zn2+, Co2+, Cu2+, or Ni2 directly to the assay
mixture. From these results, we concluded that Bvu0220 is not dependent
on the binding of divalent cations for catalytic activity and that
the function of this enzyme is to hydrolyze lactones of acid sugars.
The kinetic constants for the hydrolysis of lactones 1 through 5 by Bvu0220 are presented in Table 1. l-Glucono-1,5-lactone was the best of
these substrates with a value of kcat/Km of 8 × 104 M–1 s–1.
Scheme 1
Table 1
Catalytic Constants
for Bvu0220, Bvu0219,
and Bvu0222
enzyme
substrate
kcat (s–1)
Km (μM)
kcat/Km (M–1 s–1)
Bvu0220a
l-glucono-1,5-lactone (1)
38 ± 1
500 ± 40
(8.0 ± 0.5) × 104
l-galactono-1,4-lactone (2)
34 ± 3
1950 ± 310
(1.8 ± 0.2) × 104
l-fucono-1,4-lactone (3)
23 ± 2
1320 ± 210
(1.8 ± 0.2) × 104
4-deoxy-l-fucono-1,5-lactone (12)
23 ± 1
32 ± 4
(7.2 ± 0.8) × 105
l-xylono-1,4-lactone (5)
5.3 ± 0.4
5500 ± 550
(9.3 ± 0.4) × 102
d-arabino-1,4-lactone (4)
1.0 ± 0.1
1700 ± 140
(5.0 ± 0.3) × 102
Bvu0219b
l-galactose (6)
21 ± 1
109 ± 4
(2.0 ± 0.1) × 105
l-glucose (7)
NDe
ND
(2.0 ± 0.2) × 102
d-arabinose (8)
ND
ND
(7.1 ± 0.1) × 101
l-fucose (9)
ND
ND
(3.4 ± 0.1) × 101
d-altrose (10)
ND
ND
(1.3 ± 0.1) × 101
NADP+ (with l-galactose)
22 ± 1
3.0 ± 0.5
(8.0 ± 0.4) × 106
NAD+ (with l-galactose)
25 ± 1
3000 ± 200
(9.0 ± 0.1) × 103
Bvu0222c
l-galactonate (13)
0.6 ± 0.1
36 ± 3
(1.7 ± 0.1) × 10 4
l-gluconate (14)
0.2 ± 0.1
510 ± 70
(3.9 ± 0.6) × 102
l-fuconate
ND
ND
(2.0 ± 0.1) × 102
d-tagaturonate (15)d
90 ± 0.6
550 ± 50
(1.6 ± 0.2) × 105
Bromothymol blue pH indicator assay
and MOPS (pH 7.1).
50 mM
BICINE buffer (pH 8.0), varying
concentrations of substrate and 0.5 mM NADP+.
50 mM phosphate buffer (pH 7.5)
varying concentrations of substrate and 0.5 mM NAD+.
50 mM phosphate buffer (pH 7.0)
varying concentrations and 1.0 mM NADH.
ND, did not determine because the
enzyme was not saturated at the highest substrate concentration employed.
Bromothymol blue pH indicator assay
and MOPS (pH 7.1).50 mM
BICINE buffer (pH 8.0), varying
concentrations of substrate and 0.5 mM NADP+.50 mM phosphate buffer (pH 7.5)
varying concentrations of substrate and 0.5 mM NAD+.50 mM phosphate buffer (pH 7.0)
varying concentrations and 1.0 mM NADH.ND, did not determine because the
enzyme was not saturated at the highest substrate concentration employed.
Functional Characterization
of Bvu0219
The gene for
Bvu0219 is adjacent to the gene for Bvu0220. This protein is currently
annotated as an “oxidoreductase” from cog0667, and the
functionally characterized enzymes from cog0667 include dehydrogenases
for the oxidation of l-fucose and d-arabinose. Bvu0219
also shares significant sequence identity (42–46%) to l-galactose dehydrogenase from two plants species, ornamental tobacco
(Nicotiana langsdorffii x Nicotiana sandera) and Arabidopsis thaliana. It was therefore predicted
that Bvu0219 would oxidize a monosaccharide to the corresponding lactone
and that the product of the reaction catalyzed by Bvu0219 would be
the physiological substrate for the lactonase activity of Bvu0220.
Bvu0219 was screened against a substrate library of 23 commercially
available pentose and hexose sugars by following the reduction of
NADP+ at 340 nm. Catalytic activity was observed for l-galactose (6), l-glucose (7), d-arabinose (8), l-fucose (9), and d-altrose (10) using NADP+ as the oxidant. The structures of these compounds are presented
in Scheme 2, and the kinetic constants are
presented in Table 1. l-Galactose
is the best substrate identified for Bvu0219 with a value of kcat/Km of 2 ×
105 M–1 s–1. The rate
constants for the other substrates are reduced by more than 3 orders
of magnitude.
Scheme 2
Identification of the Initial Reaction Product
of Bvu0219
Previous reports have suggested that the reaction
product for the
oxidation of l-galactose is l-galactono-1,4-lactone
(2) rather than l-galactose-1,5-lactone (11).[23−26]l-Galactose dehydrogenase activity has previously been
reported in plant species, where this reaction is important for the
biosynthesis of ascorbic acid.[27] It is
likely that the initial product is actually l-galactono-1,5-lactone
(11) since the pyranose form of l-galactose
is the predominant configuration in solution.[28] However, l-galactono-1,5-lactone (11) has
not been observed previously due to its chemical instability and nonenzymatic
rearrangement to l-galactono-1,4-lactone (2).
We therefore attempted to determine the structure of the initial oxidation
product in the reaction catalyzed by Bvu0219 by NMR spectroscopy using
[13C-1]-l-galactose. The enzymatic reaction was
conducted at pH 5.8 to help stabilize the initial oxidation product.[13]The 13C NMR spectrum for [13C-1]-l-galactose is presented in Figure 3A. The four resonances correspond to C-1 for each
of the four anomers of l-galactose. The chemical shifts of
the α- and β-pyranose forms are observed at 96.5 and 92.3
ppm, respectively. The α- and β-furanose anomers are found
at 95.0 and 101.0 ppm, respectively. The anomeric ratios are consistent
with data reported by Angyal and Pickles: β-pyranose (64%),
α-pyranose (29%), β-furanose (4%), and α-pyranose
(3%).[28] After Bvu0219 is added to a mixture
of NAD+ and [13C-1]-l-galactose at
pH 5.8, a new resonance begins to appear at 174.1 ppm. Over a period
of 15 min, the resonance at 174.1 ppm diminishes and a new resonance
begins to appear at 175.9 (Figure 3B). The
resonance at 175.9 ppm is assigned to that of l-galactono-1,4-lactone
(2) based on a direct comparison with the chemically
synthesized material. The transient resonance that appears at 174.1
ppm is that of l-galactono-1,5-lactone (11).
Therefore, the initial product of the reaction catalyzed by Bvu0220
is l-galactono-1,5-lactone (11). This unstable
product is nonenzymatically converted to l-galactono-1,4-lactone
(2) with an estimated rate constant of ∼0.2 min–1 at pH 7.0.
Figure 3
13C NMR spectra for [13C-1]-l-galactose
and the nonenzymatic transformation of l-galactono-1,5-lactone
(11) to l-galactono-1,4-lactone (2). (A) The four 13C-resonances correspond to the anomeric
carbon for each of the four anomers of l-galactose (6). Chemical shifts are as follows: α-pyranose (92.3
ppm), β-pyranose (96.5 ppm), α-furanose (95.0 ppm), and
β-furanose (101.0 ppm). (B) (I) The carbonyl region of l-galactose (6) is presented at time 0. (II) Bvu0219
added with l-galactono-1,5-lactone (11) as the
predominate species (174.1 ppm). The enzyme was removed and the pH
adjusted to 7. (III) The resonance signal for l-galactono-1,5-lactone
decreases as the signal for l-galactono-1,4-lactone (175.9
ppm) increases in the absence of enzyme. (IV) After 10 min the predominant
species is l-galactono-1,4-lactone (2). (V)
After 15 min all of the l-galactono-1,5-lactone has been
converted to l-galactono-1,4-lactone.
13C NMR spectra for [13C-1]-l-galactose
and the nonenzymatic transformation of l-galactono-1,5-lactone
(11) to l-galactono-1,4-lactone (2). (A) The four 13C-resonances correspond to the anomeric
carbon for each of the four anomers of l-galactose (6). Chemical shifts are as follows: α-pyranose (92.3
ppm), β-pyranose (96.5 ppm), α-furanose (95.0 ppm), and
β-furanose (101.0 ppm). (B) (I) The carbonyl region of l-galactose (6) is presented at time 0. (II) Bvu0219
added with l-galactono-1,5-lactone (11) as the
predominate species (174.1 ppm). The enzyme was removed and the pH
adjusted to 7. (III) The resonance signal for l-galactono-1,5-lactone
decreases as the signal for l-galactono-1,4-lactone (175.9
ppm) increases in the absence of enzyme. (IV) After 10 min the predominant
species is l-galactono-1,4-lactone (2). (V)
After 15 min all of the l-galactono-1,5-lactone has been
converted to l-galactono-1,4-lactone.
Identification of the Physiological Substrate for Bvu0220
The best substrate for Bvu0219 is clearly l-galactose,
but the best substrate for Bvu0220 is not the corresponding l-galactono-1,4-lactone (2). To determine if the best
substrate for Bvu0220 is actually the chemically unstable l-galactono-1,5-lactone (11), Bvu0219 was incubated with
[13C-1]-l-galactose and NAD+ at pH
5.8, and the reaction was monitored via 13C NMR spectroscopy
(Figure 4). After 15 min the resonances for
both lactones were approximately equal in intensity (Figure 4C). Bvu0220 was then added to the reaction mixture,
and the 13C NMR spectrum was monitored as a function of
time (Figure 4D). The resonance at 174.1 ppm,
corresponding to l-galactono-1,5-lactone (11), disappeared rapidly, and a new resonance appeared at 179.4 ppm,
which corresponds to the hydrolysis product, l-galactonate
(13). Similarly, when both enzymes (Bvu0219 and Bvu0220)
were added to a mixture of NAD+ and [13C-1]-l-galactose at pH 5.8, the resonance for l-galactono-1,4-lactone
(2) was not observed. The only resonance that was observed
was that for l-galactonate (13) at 179.4 ppm
(data not shown). We estimate by NMR spectroscopy that Bvu0220 hydrolyzes l-galactono-1,5-lactone (11) 300 times faster than l-galactono-1,4-lactone (2). Therefore, the likely
physiological substrate for Bvu0220 is l-galactono-1,5-lactone
(11) that results from the oxidation of l-galactose
(6) by Bvu0219 (Scheme 3).
Figure 4
13C NMR analysis of the reactions catalyzed by Bvu0219
and Bvu0220 utilizing [13C-1]-l-galactose as substrate.
(A) The carbonyl region of [13C-1]-l-galactose.
(B) The predominate species after the addition of Bvu0219 is l-galactono-1,5-lactone (174.1 ppm, red asterisk). (C) 10 min after
the addition of Bvu0219 the resonances corresponding to C-1 of l-galactono-1,5-lactone and l-galactono-1,4-lactone
(175.9 ppm, blue asterisk) are approximately equal. At this point
Bvu0220 was added to the reaction mixture. (D) 5 min after the addition
of Bvu0220, l-galactono-1,5-lactone has disappeared. A new
resonance with a chemical shift of 179.4 ppm appeared, which corresponds
to l-galactonate (green asterisk).
Scheme 3
13C NMR analysis of the reactions catalyzed by Bvu0219
and Bvu0220 utilizing [13C-1]-l-galactose as substrate.
(A) The carbonyl region of [13C-1]-l-galactose.
(B) The predominate species after the addition of Bvu0219 is l-galactono-1,5-lactone (174.1 ppm, red asterisk). (C) 10 min after
the addition of Bvu0219 the resonances corresponding to C-1 of l-galactono-1,5-lactone and l-galactono-1,4-lactone
(175.9 ppm, blue asterisk) are approximately equal. At this point
Bvu0220 was added to the reaction mixture. (D) 5 min after the addition
of Bvu0220, l-galactono-1,5-lactone has disappeared. A new
resonance with a chemical shift of 179.4 ppm appeared, which corresponds
to l-galactonate (green asterisk).To further investigate the preferential hydrolysis of
1,5-lactones
by Bvu0220, 4-deoxy-l-fucono-1,5-lactone (12) was synthesized enzymatically using l-fucose (9) and l-fucose dehydrogenase (FucD). This lactone is missing
the hydroxyl group at C-4 and therefore cannot rearrange to a 1,4-lactone.
The enzymatic product, 4-deoxy-l-fucono-1,5-lactone (12), was confirmed through 1H NMR spectroscopy
(data not shown). The 4-deoxy-l-fucono-1,5-lactone (12) was isolated, and the kinetic constants for the hydrolysis
by Bvu0220 were determined at pH 7.1. The value of kcat/Km for the hydrolysis
of 4-deoxy-l-fucono-1,5-lactone (12) is approximately
an order of magnitude greater than that for l-glucono-1,5-lactone
(1) and almost 2 orders of magnitude greater than for l-fucono-1,4-lactone (3).
Functional Characterization
of Bvu0222
The gene for
Bvu0222 is adjacent to Bvu0219 and Bvu0220, and this protein is currently
annotated as a Zn2+-dependent oxidoreductase from cog1063.
Functionally characterized members of cog1063 include l-threonine
dehydrogenase, l-idonate dehydrogenase, and sorbitol dehydrogenase.
The closest homologue with an experimentally determined catalytic
function is LgnH (gi|425703034, UniProt K7ZKU8) from Paracoccus
sp. 43P, with a sequence identity of 46%. The gene for LgnH
is contained in an operon for the catabolism of the rare sugar l-glucose. LgnH has been reported to oxidize l-galactonate
(13) and l-gluconate (14) at approximately
the same rate.[29] Bvu0222 catalyzes the
oxidation of l-galactonate (13), l-gluconate
(14), and l-fuconate (15); however,
the value of kcat/Km for the oxidation of l-galactonate (13) is 2 orders of magnitude greater than the oxidation of l-gluconate (14) (Table 1). d-Arabinonate and l-xylonate were also assayed for
activity; however, no oxidation was detected. The oxidation product
of l-galactonate (13) by Bvu0222 was confirmed
to be d-tagaturonate (16). d-Tagaturonate
(16) was prepared by the isomerization of d-galacturonate
(17) by the enzyme uronate isomerase.[17] The kinetic constants for the reduction of d-tagaturonate
(16) are presented in Table 1,
and the structures of compounds 13 through 17 are presented in Scheme 4.
Scheme 4
Discussion
Novel Metabolic
Pathway for l-Galactose
The
three enzymes from B. vulgatus examined in this study
function to form a previously unknown pathway for the metabolism of
l-galactose (6). The first enzyme in the pathway,
Bvu0219, preferentially oxidizes l-galactose (6) to l-galactono-1,5-lactone (11) using NADP+ as the oxidant with a kcat/Km of >105 M–1 s–1. The next best substrate, l-glucose
(7), is oxidized 3 orders of magnitude less efficiently.
The l-galactono-1,5-lactone (11) is chemically
unstable
and rapidly isomerizes to l-galactono-1,4-lactone (2). We suggest that Bvu0219 be called l-galactose
dehydrogenase and that the corresponding gene be designated as lgaA. In the subsequent step, Bvu0220 hydrolyzes l-galactono-1,5-lactone (11) to l-galactonate
(13). The substrate specificity of Bvu0220 is broader
than that of Bvu0219, but the two 1,5-lactones (l-glucono-1,5-lactone
(1) and 4-deoxy-l-fucono-1,5-lactone (12)) are hydrolyzed about an order of magnitude faster than
any of the 1,4-lactones, including l-galactono-1,4-lactone
(2). However, the clear preference for the hydrolysis
of l-galactono-1,5-lactone (11) by Bvu0220 was
demonstrated with NMR spectroscopy using Bvu0219 to rapidly oxidize
[13C-1]-l-galactose to l-galactono-1,5-lactone
(11). We suggest that this enzyme be named l-galactono-1,5-lactonase and the corresponding gene be denoted as lgaB. In the final step, Bvu0222 oxidizes l-galactonate
(13) at C-5 to form d-tagaturonate (16). We propose that Bvu0222 be called l-galactonate dehydrogenase
and the gene that encodes this protein be labeled as lgaC. The overall reaction pathway is presented in Scheme 5.
Scheme 5
Metabolism of d-Tagaturonate
What is the probable
fate of d-tagaturonate? In bacteria such as E. coli, the keto group at C2 of d-tagaturonate (16) is reduced to form d-altronate (18) by UxaB,
followed by the UxaA catalyzed dehydration of d-altronate
to 2-keto-3-deoxygluconate (KDG) (19). This product is
then phosphorylated by KDG kinase (KdgK) to form 2-keto-3-deoxygluconate-6-phosphate
(20) and then cleaved by an aldolase (KdgA) to generate
pyruvate and d-glyceraldehyde-3-phosphate. These steps are
illustrated in Scheme 6. In B. vulgatus, the four proteins needed for the metabolism of d-tagaturonate
can be identified with high probability based on their sequence similarity
to the experimentally characterized enzymes from E. coli. Bvu3075 is 46% identical to d-tagaturonate reductase (UxaB),
Bvu3053 is 46% identical to dD-altronate dehydratase (UxaA),
Bvu3055 is 25% identical to KdgK and Bvu3056 is 33% identical to KdgA.
Therefore, it appears that B. vulgatus can metabolize l-galactose to pyruvate and d-glyceraldehyde by a pathway
that consists of seven steps.
Scheme 6
Bvu0219 and Other Enzymes
in cog0667
Experimentally
verified members of cog0667 include the following enzymes: l-glyceraldehyde phosphate reductase, d-arabinose dehydrogenase, l-fucose dehydrogenase, and l-galactose dehydrogenase.[24,30−32] As previously mentioned, Bvu0219 shares approximately
42% sequence identity with l-galactose dehydrogenase from A. thaliana. We have predicted that at least 35 proteins
in the NCBI protein database share the same substrate profile as Bvu0219.
This prediction is based on significant sequence identity (>40%)
and
if the protein of interest is encoded adjacent to homologues of Bvu0220
and Bvu0222.
Bvu0220 and Other Enzymes in cog3618
We have previously
determined the three-dimensional structures of two enzymes from cog3618,
FucL (PDB id: 4DNM) and LigI (PDB id: 4D8L). LigI catalyzes the hydrolysis of 2-pyrone-4,6-dicarboxylate in
the degradation of lignin. Bvu0220 shares approximately 28% and 15%
sequence identity with FucL and LigI, respectively. Despite the relatively
low sequence identity to other members of cog3618, Bvu0220 conserves
important catalytic active site residues (Figure 5). These include the HxH motif from β-strand 1, histidine
from β-strand 6, arginine from β-strand 4, and aspartate
from β-strand 8. A total of 12 proteins in the SSN for cog3618
(Figure 2) are predicted to share the same
substrate profile as Bvu0220. In addition, there are approximately
35 proteins in the NCBI protein database (not represented in the SSN)
that are predicted to share this substrate profile (Table S1, Supporting Information).
Figure 5
Sequence alignment of
Bvu0220, LigI (gi|374074685), and BmulJ_04915
(l-fucono-1,5-lactonase. Residues highlighted in maroon are
conserved active site residues important for catalysis. Black arrows
represent β-strands that make up the (β/α)8-barrel.
Sequence alignment of
Bvu0220, LigI (gi|374074685), and BmulJ_04915
(l-fucono-1,5-lactonase. Residues highlighted in maroon are
conserved active site residues important for catalysis. Black arrows
represent β-strands that make up the (β/α)8-barrel.
Bvu0222 and Other Enzymes
in cog1063
Cog1063 contains
primarily oxidoreductase enzymes including d-glucose dehydrogenase, l-idonate dehydrogenase, l-arabinitol dehydrogenase, l-threonine dehydrogenase, d-xylulose dehydrogenase,
sorbitol dehydrogenase, and l-gluconate dehydrogenase.[33−39] As mentioned, the closest homologue of verified function is l-gluconate dehydrogenase (LgnH) from Paracoccus sp.
43P, which shares 46% sequence identity. LgnH was reported
to oxidize both l-gluconate and l-galactonate with kcat/Km values >
104 M–1 s–1. Conversely,
Bvu0222 oxidizes l-galactonate with a kcat/Km value approximately 2 orders
of magnitude greater than that for l-gluconate. As many as
35 protein sequences have been identified within the NCBI database
and predicted to share the same substrate profile as Bvu0222 (Table S3, Supporting Information).
Metabolic Pathways
for l-Galactose
l-Galactose is most notable
as an intermediate in the biosynthesis
of ascorbic acid in higher plants. For ascorbic acid biosynthesis, l-galactose is produced through a series of reactions, beginning
with d-glucose-6-phosphate, which is ultimately converted
to GDP-d-mannose. GDP-d-mannose is then transformed
to GDP-l-galactose through epimerization at C3 and C4 of
the sugar moiety. Phosphorolysis of this product yields GDP and l-galactose-1-phosphate, which is subsequently dephosphorylated
to l-galactose.[27]l-Galactose
was first isolated from flaxseed oil in 1903, and it was later identified
to be a component of certain pectins.[41] Since then, l-galactose has been identified as a component
of agar, galactagen, agaropectin, and rhamnogalacturonan II (RG-II).[41] RG-II is a structural component of plant cell
walls, which makes up approximately 10% of the total pectin in the
biosphere. 6-Deoxy-l-galactose (l-fucose) is a known
component of RG-II. It has recently been discovered that l-galactose is naturally substituted for l-fucose in RG-I.[41] For example, 24% of RG-II in red wine and 14%
in carrots (Daucus carota) is substituted with l-galactose.This is the first pathway described for the
transformation of l-galactose to d-tagaturonate.
Bioinformatics analysis reveal this pathway to be prevalent in many Bacteroides species as well as in other gut microbes, such
as Paraprevotella clara, Parabacteroides
goldsteinii, Paraprevotella xylaniphila,
and Parabacteroides sp. CAG:409. In addition to gut
microbes, the pathway is found in Prevotella paludivivens and Proteiniphilum acetatigenes isolated from a
rice field in Japan and brewery wastewater, respectively.
Discovery of
Novel Metabolic Pathways
In this investigation
we identified and characterized a novel pathway for the metabolism
of l-galactose in bacteria that reside within the human gut.
This discovery was initiated by the comprehensive application of bioinformatics
and sequence similarity networks for the identification of enzymes
of unknown function that are homologues to previously characterized
lactonases from the amidohydrolase superfamily in cog3618. The gene
for one of these enzymes of unknown function, Bvu0220, was found clustered
in an apparent operon for carbohydrate metabolism. The catalytic properties
of this enzyme were elucidated through the utilization of a small,
but focused, library of potential sugar lactones. Sequence similarity
networks and small physical libraries of potential substrates were
strategically utilized to identify substrate profiles for two other
enzymes of unknown function in this putative operon for the metabolism
of l-galactose. Similar strategies can be readily applied
in a broader effort to annotate previously uncharacterized metabolic
pathways.
Authors: Dao Feng Xiang; Peter Kolb; Alexander A Fedorov; Chengfu Xu; Elena V Fedorov; Tamari Narindoshivili; Howard J Williams; Brian K Shoichet; Steven C Almo; Frank M Raushel Journal: Biochemistry Date: 2012-02-15 Impact factor: 3.162
Authors: Daniel S Hitchcock; Alexander A Fedorov; Elena V Fedorov; Lawrence J Dangott; Steven C Almo; Frank M Raushel Journal: Biochemistry Date: 2011-06-07 Impact factor: 3.162
Authors: Richard S Hall; Rakhi Agarwal; Daniel Hitchcock; J Michael Sauder; Stephen K Burley; Subramanyam Swaminathan; Frank M Raushel Journal: Biochemistry Date: 2010-05-25 Impact factor: 3.162
Authors: Richard S Hall; Alexander A Fedorov; Ricardo Marti-Arbona; Elena V Fedorov; Peter Kolb; J Michael Sauder; Stephen K Burley; Brian K Shoichet; Steven C Almo; Frank M Raushel Journal: J Am Chem Soc Date: 2010-02-17 Impact factor: 15.419
Authors: Merlin Eric Hobbs; Matthew Vetting; Howard J Williams; Tamari Narindoshvili; Devon M Kebodeaux; Brandan Hillerich; Ronald D Seidel; Steven C Almo; Frank M Raushel Journal: Biochemistry Date: 2012-12-20 Impact factor: 3.162
Authors: Pavel Savitsky; James Bray; Christopher D O Cooper; Brian D Marsden; Pravin Mahajan; Nicola A Burgess-Brown; Opher Gileadi Journal: J Struct Biol Date: 2010-06-10 Impact factor: 2.867
Authors: Craig S Robb; Joanne K Hobbs; Benjamin Pluvinage; Greta Reintjes; Leeann Klassen; Stephanie Monteith; Greta Giljan; Carolyn Amundsen; Chelsea Vickers; Andrew G Hettle; Rory Hills; Xiaohui Xing; Tony Montina; Wesley F Zandberg; D Wade Abbott; Alisdair B Boraston Journal: Nat Chem Biol Date: 2022-03-14 Impact factor: 16.174
Authors: Leila B Giron; Ceylan E Tanes; Mariane H Schleimann; Phillip A Engen; Lisa M Mattei; Alitzel Anzurez; Mohammad Damra; Huanjia Zhang; Kyle Bittinger; Frederic Bushman; Andrew Kossenkov; Paul W Denton; Hiroaki Tateno; Ali Keshavarzian; Alan L Landay; Mohamed Abdel-Mohsen Journal: Mucosal Immunol Date: 2020-03-09 Impact factor: 7.313
Authors: Jason W Arnold; Joshua B Simpson; Jeffery Roach; Jose M Bruno-Barcena; M Andrea Azcarate-Peril Journal: Nutrients Date: 2018-10-16 Impact factor: 5.717